Collision Detection Using Spatial Curves and Voxel Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for collision detection between moving objects in machine tools, such as the sweep method and time division method, face challenges with high computational costs and potential missed collisions due to coarse or fine division intervals, especially when handling three-dimensional shapes, and require significant memory for processing.
Innovation Solution
A collision determination device that uses a voxel model for one object and a boundary representation model for the other, employing determination test points, spatial curves, and distance computation to assess intersection and collision possibility, reducing computational costs and memory usage while ensuring reliable detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the sweep method is used to detect collision by creating a sweep shape of the moving object along its trajectory, then the collision detection can be performed as a continuous event, but the computational cost increases significantly and the method is not readily applicable to arbitrary shapes and movements
Solution Approach 1:
The patent segments the continuous movement trajectory into discrete spatial segments based on the moving object's position changes. Instead of creating a complete sweep shape for the entire trajectory, the method divides the space into regions and determines collision by checking which regions the moving object passes through, significantly reducing computational complexity while maintaining detection accuracy.
Solution Approach 2:
The patent transforms the collision detection problem from three-dimensional sweep volume intersection to a two-dimensional problem by projecting the movement trajectory onto a plane and analyzing the swept area. This dimensional reduction simplifies the computational burden while preserving the essential collision detection capability.
2Productivity
If the time division method is used to determine intersection at every instant by dividing the duration of movement, then the computational cost per determination is reduced, but finer division intervals increase the number of determinations and coarser intervals cause missed collisions
Solution Approach 1:
The patent dynamically adjusts the spatial segmentation based on the moving object's velocity and trajectory characteristics. Regions with higher movement density are divided into smaller segments while regions with lower activity use larger segments, optimizing the balance between detection accuracy and computational efficiency without requiring uniform fine division throughout the entire trajectory.
Solution Approach 2:
The patent performs preliminary analysis of the movement trajectory to identify key positions and potential collision zones before conducting the actual collision detection. By pre-processing the trajectory data and marking critical regions, the method avoids unnecessary intersection determinations in safe zones while ensuring thorough checking in high-risk areas.
3Adaptability or versatility
If space grid model such as voxel model is used to represent the shape of objects for collision determination, then the method can handle arbitrary shapes including works with complicated shapes during processing, but the amount of memory required increases considerably
Solution Approach 1:
The patent applies local quality by representing only the necessary portions of objects in voxel format rather than entire objects. The moving object is represented using simplified geometric primitives while the stationary object uses voxel representation only in regions where collision detection is required. This selective application of detailed representation reduces memory usage while maintaining detection accuracy for potential collision zones.
Solution Approach 2:
The patent segments the spatial domain into a hierarchical structure where only relevant regions are represented in detailed voxel format. By dividing the workspace into coarse and fine regions, and using voxel representation only for fine regions near potential collision points, the method achieves both shape flexibility and memory efficiency.
Data Source
AI summary
A collision determination device includes a target cell designation unit that designates a target cell for a voxel model which represents the shape of a first object. A determination test point generation unit generates determination test points for the target cell. A spatial curve creation unit creates a spatial curve as the trajectory curve of a determination test point. A curve intersecting determination unit determines whether or not each boundary element of a boundary representation model representing the shape of a second object is intersecting the spatial curve. A distance computation unit computes the closest distance from the spatial curve to the boundary surface of the boundary representation model. A collision determination unit determines, based on the determination result and the computation result, whether there is a possibility of collision between both objects.